We propose to 1) Determine the origin of SV40-like sequences in human tumors; 2) Investigate if these viral sequences contribute to carcinogenesis; 3) Study if the expression of viral sequences can be used to develop new diagnostic approaches for mesothelioma and osteosarcoma patients. We will investigate the possible role of SV40-like DNA sequences in the development of human mesotheliomas and osteosarcomas. We will sequence, by PCR, the entire viral genome present in mesotheliomas and in osteosarcomas to determine how much of the viral genome is present in these human tumors and whether it is identical to the SV40 genome or mutated. We will attempt to isolate and sequence the SV40-like viruses from human mesotheliomas and osteosarcomas by transfecting DNAs from these tumors into permissive CV1 monkey cells. Next, we will investigate human biopsies by PCR, Northern and Western blot experiments, and immunohistochemistry, to determine if the SV40-like antigen (Tag) produced in human mesotheliomas binds and inactivates the cellular tumor suppressor proteins known to bind Tag in tissue culture: p53, Rb, pRb2, p107, p300. In addition, we will investigate by immunoperoxidase and Western blot experiments if the high level of c-fos expression we found in poor-risk osteosarcoma patients are induced by SV40-like in these tumors. Because SV40 small t antigen (tag) induces -fos in monkey cells in culture, we will transfect tag into human osteosarcoma cells in culture to test if tag can induce c-fos expression also in these cells. We will also investigate if the presence of SV40-like sequences in malignant mesothelioma cells can represent a useful marker for the pathologist for confirming, using PCR amplification, the often difficult morphologic diagnosis of mesothelioma.